On 13/11/2016 09:27, Norman Lynagh wrote:
Ian wrote:

I've always thought it a bit daft when recording 09-09 that particularly
In winter your 09 obs of for example -2C is quite often your min for the
next 09-09 period if there is a change to a milder airmass.
2 air frost days for the price of 1.

Ian Raunds E Northants

The same sort of thing happens no matter what recording period you use. For
example, if there is a very short period of air frost from 2330 till 0030 that
would only count as 1 air frost day if using the 09-09 recording period but it
would count as 2 air frost days if you use the 00-24 recording period. Indeed a
single night of air frost lasting, say, from 2000 till 0800 would count as 1
air frost day in the 0900-0900 recording system but would count as 2 days in
the 0000-2400 system, even though there was actually only one night of air
frost. No system is perfect. The important thing is to have a system and stick
to it so that there is internal consistency withing your own set of
observations. Strictly speaking, comparison with surrounding "official"
stations is valid only if you use the 0900-0900 GMT recording period.


On the Winter 1947 website I wrote

'Finally, from other sources, the minimum overnight temperature at
Writtle (Essex) on the 28th/29th January was a rather chilly -5.1°F
(-20.6°C) recorded at 0900 on the 29th.

That 0900 reading has been assigned to 28th and/or 29th depending where
you look.

--
George in Swanston, Edinburgh, 580'asl
www.swanstonweather.co.uk
www.eppingweather.co.uk
www.winter1947.co.uk